Abstract Gram-negative bacterial infections account for more than half a million severe sepsis cases every year in the United States, where a major pathogenic factor is lipopolysaccharides (LPS) and their corresponding toxic component, lipid A. Minimizing serum lipid A level was expected to be an effective therapy for managing severe sepsis. This hypothesis has been validated in clinical studies. For instance, hemoperfusion methods using immobilized polymyxin B (PB) to remove circulating LPS have proven beneficial. However, common LPS- neutralizing agents, such as PB, are highly toxic, therefore have rather restricted therapeutic applications. In addition, the repertoire of LPS-binding agents is limited by their naturally occurring precursors. On the other hand, due to the difficulties in targeting the lipid A moiety, de novo discovery of LPS-neutralizing compounds remains challenging. Herein, a modular screening approach will be implemented to identify lipid A-specific binders form cyclic peptide libraries. A rhodamine-phenylboronic acid conjugate will be used to stain the polysaccharide part in LPS, which allows differentiating polysaccharide-specific binders from lipid A-specific ones. Three orthogonal methods will be employed to validate the affinity and specificity of identified peptides. Further in vitro validations will be performed on bone marrow-derived macrophages and human peripheral blood mononuclear cells to test the bioactivity. These cells will be challenged with LPS in conjunction with the peptides and their cytokine production will be evaluated. Lower cytokine levels indicate better LPS neutralization effects. Ultimately, the goal is to identify pan-LPS reactive peptides that exhibit high LPS-neutralization capability as well as low cytotoxicity.